Emission of pesticides into the air

Transcription

1 during and after application to a potato crop F. van den Berg 1, H.J. Holterman 2 and M. Leistra 1 1 Alterra, Centre for Water and Climate 2 Plant Research International Contents Introduction Description of field experiments Description of IDEFICS and parameterisation Description of PEARL and parameterisation Comparison of computations with measurements Discussion and conclusions 1

2 Introduction Substantial emissions can occur during and after application Comparatively high volatilisation rates after crop spraying Most pesticides are applied to the crop Use of models to assess exposure Field experiment at Slootdorp Emission during application spraying of tracer (BSF) on potato crop passive and active sampling measurements of deposition and airborne drift Emission after application spraying of fenpropimorph and chlorpyrifos measurements of emission fluxes with different methods residues on plant surfaces 2

3 Emission during application: experimental set-up wind sampling poles 27 m Swath 3 Swath 2 Swath 1 ground deposits Field experiment at Slootdorp (NL) Application to potato crop on 25 June 2002 Meteorological data obtained by measurements in the field 3

4 Post-application emission: spraying of pesticides Volatilisation measurements at Slootdorp. Application at 25 June

5 Transport pathways during application sprayer nozzle large drops wind small drops airborne spray crop ditch surface water IDEFICS spray drift model Purpose Downwind ground deposits (0 15m) Airborne spray distribution Developed for conventional field sprayer Basic concepts Particle tracking model (mixed 2D/3D) Equations of motion (deterministic) Turbulent air flow (stochastic) Evaporating droplets ( solid core ) 5

6 Deposition on the ground as computed with IDEFICS ground deposits [%appl.dose] measured: swath 1 measured: swaths 2+3 simulation: swath 1 simulation: swath 1, droplets simulation: swaths distance downwind from crop edge [m] Conclusion: deposits computed corresponds with those measured Airborne drift from swath 1 at downwind distance of 5.5 m airborne drift [%appl dose] swath 1 measured: passive measured: passive; eff 35% measured: active simulation; droplets simulation; droplets+dust height above ground level [m] Conclusion: model overestimates drift; shape is different 6

7 Airborne drift from swaths 2+3 at downwind distance of 5.5 m airborne drift [%appl dose] swath 2+3 measured: passive measured: passive; eff 35% measured: active simulation: droplets simulation: droplets+dust height above ground level [m] Conclusion: model underestimates drift Averaged airborne emission from treated area at 5.5 m downwind swath m 1) Flux 2) [mg.m 2.hr 1 ] Measurement [%applied] Simulation [%applied] ) Estimate for a 100 m wide field, assuming a 4 th swath identical to the 3 rd. 2) assumed: 300 l/ha; 5 mg/ml a.i.; 1.8 m/s driving speed; 1 ha field Conclusion: significant airborne fraction downwind from field 7

8 Some major problems in research on emission during application Sampling techniques show high variation Collection efficiency of samplers not very accurate Accurate estimation of droplet paths near the nozzle probably is critical for emission into the air Processes affecting the fate of pesticides on plants Photo-transformation Volatilisation Droplet Penetration Plant leaf Wash-off by rain 8

9 Description of processes on plants in PEARL Volatilisation: laminar air boundary layer Penetration, phototransformation and wash-off: firstorder processes Rate coefficient of phototransformation function of actual solar radiation Derivation of input for pesticides Property Vapour pressure Half life for penetration Half life for photo transformation Laminar boundary layer Chlorpyrifos 2.7 mpa at 25 C 3 d 3 d 0.2 mm Fenpropimorph 2.2 mpa at 20 C 0.13 d 0.13 d 0.2 mm 9

10 Volatilisation of chlorpyrifos from potatoes computed with PEARL Conclusion: high initial volatilisation losses; diurnal pattern Mass balance of chlorpyrifos on potatoes computed with PEARL Conclusion: volatilisation dominant process 10

11 Volatilisation of fenpropimorph from potatoes computed with PEARL Conclusion: Fast decline in volatilisation rate Mass balance of fenpropimorph on potatoes computed with PEARL Conclusion: Low volatilisation, other processes dominate 11

12 Vapour pressure is crucial parameter in modelling volatilisation from plants Critical evaluation of available data needed as values may differ strongly Value may not correspond to volatilisation behaviour of pesticide (relative to others) Measurement should be made following latest OECD Guideline Estimation methods may be needed for checking Some major problems in research on emission after application Reliability of the input data, e.g. vapour pressure Lack of direct input data, e.g. on penetration, wash off, photo transformation Not present as pure compound; effect of substances e.g. in formulation Very complex geometry of canopy, deposit, etc. Complexity and variability of weather conditions 12

13 Further research Improvement of sampling techniques for airborne drift Description of airborne drift for orchard spraying Experiments to derive rate coefficients for processes competing with volatilisation on plant surfaces Description of the distribution of pesticide deposit in the canopy Description of atmospheric resistances to volatilisation Conclusions Considerable progress in modelling airborne drift and volatilisation in past 4 years Emission into the air during application as well as after application can be measured The effect of environmental conditions, substance properties and application characteristics on the total emission into the air can be estimated using IDEFICS and PEARL 13

14 Thank you for your attention 14